Web Data Management - Forsiden - Universitetet i Oslo · Web Data Management ... belonging to a...
Transcript of Web Data Management - Forsiden - Universitetet i Oslo · Web Data Management ... belonging to a...
1
Web Data Management Vera Goebel
Department of Informatics, University of Oslo Fall 2015
Web Models Web Search and Querying Distributed XML Processing Why use XML with a database? -> XML-DBS architectures
Web Overview • Publicly indexable web:
– More than 25 billion static HTML pages. – Over 53 billion pages in dynamic web
• Deep web (hidden web) – Over 500 billion documents
• Most Internet users gain access to the web using search engines.
• Very dynamic – 23% of web pages change daily. – 40% of commercial pages change daily.
• Static versus dynamic web pages • Publicly indexable web (PIW) versus hidden web
(or deep web)
Properties of Web Data • Lack of a schema
– Data is at best “semi-structured” – Missing data, additional attributes, “similar” data but
not identical • Volatility
– Changes frequently – May conform to one schema now, but not later
• Scale – Does it make sense to talk about a schema for Web? – How do you capture “everything”?
• Querying difficulty – What is the user language? – What are the primitives? – Aren’t search engines or metasearch engines
sufficient?
Web Graph
• Nodes: pages; edges: hyperlinks • Properties
– Volatile – Sparse – Self-organizing – Small-world network – Power law network
Web Data Modeling • Can’t depend on a strict schema to structure the data • Data are self-descriptive
{name: {first:“Tamer”, last: “Ozsu”}, institution: “University of Waterloo”, salary: 300000}
• Usually represented as an edge-labeled graph – XML can also be modeled this way
name
institution
salary
first last
“Tamer” “Ozsu”
“UW” 300000
Web Crawling • What is a crawler? • Crawlers cannot crawl the whole Web. It should
try to visit the “most important” pages first. • Importance metrics:
– Measure the importance of a Web page – Ranking
• Static • Dynamic
• Ordering metric – How to choose the next page to crawl
Web Crawler Types • Many Web pages change frequently, so the crawler has
to revisit already crawled pages à incremental crawlers • Some search engines specialize in searching pages
belonging to a particular topic à focused crawlers • Search engines use multiple crawlers sitting on different
machines and running in parallel. It is important to coordinate these parallel crawlers to prevent overlapping à parallel crawlers
Indexing • Structure index
– Link structure • Text index
– Indexing the content – Suffix arrays, inverted index, signature files – Inverted index most common
• Difficulties of inverted index – The huge size of the Web – The rapid change makes it hard to maintain – Storage vs. performance efficiency
Web Querying • Why Web Querying?
– It is not always easy to express information requests using keywords.
– Search engines do not make use of Web topology and document structure in queries.
• Early Web Query Approaches – Structured (Similar to DBMSs): Data model +
Query Language – Semi-structured: e.g. Object Exchange Model
(OEM)
Web Querying
• Question Answering (QA) Systems – Finding answers to natural language
questions, e.g. What is Computer? – Analyze the question and try to guess what
type of information that is required. – Not only locate relevant documents but also
extract answers from them.
Approaches to Web Querying
• Search engines and metasearchers – Keyword-based – Category-based
• Semistructured data querying • Special Web query languages • Question-Answering
Semistructured Data Querying
• Basic principle: Consider Web as a collection of semistructured data and use those techniques
• Uses an edge-labeled graph model of data • Example systems & languages:
– Lore/Lorel – UnQL – StruQL
Searching The Hidden Web • Publicly indexable web (PIW)
vs. hidden web • Why is Hidden Web important?
– Size: huge amount of data – Data quality
• Challenges: – Ordinary crawlers cannot be used. – The data in hidden databases can only be accessed
through a search interface. – Usually, the underlying structure of the database is
unknown.
Searching The Hidden Web
• Crawling the Hidden Web – Submit queries to the search interface of the
database • By analyzing the search interface, trying to fill in the fields for
all possible values from a repository • By using agents that find search forms, learn to fill them, and
retrieve the result pages
– Analyze the returned result pages • Determine whether they contain results or not • Use templates to extract information
Searching The Hidden Web
• Metasearching – Database selection – Query Translation –
Result Merging – Database selection is based on Content
Summaries. – Content Summary Extraction:
• RS-Ord and RS-Lrd • Focused Probing with Database Categorization
Distributed XML • XML increasingly used for encoding web
data à data representation language – Web 2.0
• Data exchange language – Web services
• Used to encode or annotate non-web semistructured or unstructured data
• Data repository sizes growing à use distribution
19
Web Services
• Collection of protocols and standards that allows making processing requests to remote systems using a common, non-proprietary language and using common transport protocols
20
Internet Application Architecture: Today
Application messages
Browser Browser
Physical Middle Tier
Data Sources
Client Tier
ORDBMS
WEB/APP Server
Middle-Tier Application
Data Integration, Storage, Query, Management
Other Data Sources
Gateways
OLE/DB Data source
authoring tools, etc.
HTTP
HTTP
Remote messages
21
Internet Applications • Entertainment
– Games, Music, Films, Multi-person chat • Public information
– Maps, Tax return helper • Advertisement
– Interactive catalogues for products and services • Medicine
– Diagnosis, Consultation, Remote surgery • Education
– Learning-on-demand (for a degree), virtual museums, tour remote spaces
• Engineering – Collaborative design, remote parallel simulation services
• Publishing – Submit, Review, Proof-editing (text and graphics)
• Tele-communication – Conferencing
• ...
22
Web Services Architecture Principles
• Message orientation – Using only messages to communicate between services
• Protocol composability – Flexible infrastructure protocol building blocks
• Autonomous services – Endpoints can be built, deployed, managed, versioned, and
secured independently • Managed transparency
– Controlling which aspects of an endpoint are visible to external services
• Protocol-based integration – Restricting cross-application coupling to wire artifacts only
23
Web Services Core Specifications
• SOAP: Message envelope format – XML-based
• WSDL: Service interface description language – XML-based – Description of service and bindings to protocols – To generate server and client code and for configuration
• UDDI: Protocol for publishing and discovering metadata about Web services – To enable applications to find Web services at design time or
runtime • WS-Security: For secure message exchange
– How to use XML_Encryption and XML_Signature • WS-ReliableExchange: Protocol for reliable messaging
between two Web services
24
Why Use XML with a Database?
• XML as transport format – to/from database
• XML for semistructured data – schema to be stored not known at design time
• XML document archiving – retaining complete XML documents
The application domain has an impact on the choice of XDBMS
25
XML as Transport Format • When
– publishing as XML data stored in a DB, or storing data arriving as XML documents
• between an existing DB and applications • between two DBs
• Why XML – platform independence
database web service client JDBC XML
database application client XML JDBC
Publishing relational data as XML:
Storing XML data in an RDB:
26
XML for Semistructured Data
• When – application domains where
• all data formats cannot be predicted in advance + • large volumes of data
• Why – why XML: extensibility – why DB: need to store, manage, and query
data
27
XML Document Archiving
• When – retaining XML documents
• Why – processing – contracting/legislation – granularity: XML document or document
fragment is the logical unit
28
Classes of XML Documents
• Data-centric XML document: – Fairly regular structure – Fine-grained data
• Document-centric XML document: – Irregular structure – Larger grained data
The XML documents being data-centric or document-centric has an impact on the choice of XDBMS
29
Data-Centric XML Document
• Characteristics: – Discrete pieces of data – Fairly regular structure – Fine-grained data
• smallest independent unit of data is at the level of a PCDATA-only element or an attribute
– Designed for machine consumption – Not important to the application or DB that the
data is for the time being stored in an XML document
30
Example <SalesOrder Number="123"> <OrderDate>2003-07-28</OrderDate> <CustomerNumber>456</CustomerNumber> <Item Number="1"> <PartNumber>XY-47</PartNumber> <Quantity>14</Quantity> <Price>16.80</Price> </Item> <Item Number="2"> <PartNumber>B-987</PartNumber> <Quantity>6</Quantity> <Price>2.34</Price> </Item> </SalesOrder>
32
Document-Centric XML Document
• Characteristics: – Mixed content – Irregular structure – Larger grained data
• smallest independent unit of data at the level of an element with mixed content or the entire document
– Designed for human consumption – The document order (order of siblings etc.) is
significant
33
Example <Product> <Intro> The <ProductName>Turkey Wrench</ProductName> from <Developer>Full Fabrication Labs, Inc.</Developer> is <Summary>like a monkey wrench, but not as big.</Summary> </Intro> <Description> <Para>You can:</Para> <List> <Item><Link URL="Order.html">Order your own turkey wrench</Link></Item> <Item><Link URL="Wrenches.htm">Read more about wrenches</Link></Item> <Item><Link URL="Catalog.zip">Download the catalog</Link></Item> </List> <Para>The turkey wrench costs <b>just $19.99</b> and, if you order now, comes with a <b>hand-crafted shrimp hammer</b> as a bonus gift.</Para> </Description> </Product>
35
XML Document Content Types
• Element content – Content consists of child elements only
• PCDATA-only content – Content is a single piece of text
• Mixed content – Children are a mixture of text-only and
elements
36
<Product> <Intro> The <ProductName>Turkey Wrench</ProductName> from <Developer>Full Fabrication Labs, Inc.</Developer> is <Summary>like a monkey wrench, but not as big.</Summary> </Intro> <Description> <Para>You can:</Para> <List> <Item><Link URL="Order.html">Order your own turkey wrench</Link></Item> <Item><Link URL="Wrenches.htm">Read more about wrenches</Link></Item> <Item><Link URL="Catalog.zip">Download the catalog</Link></Item> </List> <Para>The turkey wrench costs <b>just $19.99</b> and, if you order now, comes with a <b>hand-crafted shrimp hammer</b> as a bonus gift.</Para> </Description> </Product>
mixed content
element content
PCDATA-only content
37
Order Can Matter! <Product> <Description> <List> <Item><Link URL="Order.html">Order your own turkey wrench</Link></Item> <Item><Link URL="Catalog.zip">Download the catalog</Link></Item> <Item><Link URL="Wrenches.htm">Read more about wrenches</Link></Item> </List> <Para>You can:</Para> <Para> and, if you order now, comes with a <b>just $19.99</b> The turkey wrench costs <b>hand-crafted shrimp hammer</b> as a bonus gift.</Para> </Description> <Intro> <Developer>Full Fabrication Labs, Inc.</Developer> from <Summary>like a monkey wrench, but not as big.</Summary> is <ProductName>Turkey Wrench</ProductName> The </Intro> </Product> ???
39
Hybrid XML Documents
• Otherwise document-centric documents, but – with parts that contain fine-grained, regularly
structured data • Otherwise data-centric documents, but
– with parts that contain large-grained, irregularly structured data
40
XDBMS - How to Achieve XML + DB Technology?
• XML + XML tools used as a ”DBMS” – Managing without a traditional DBMS
• XML-enabled DBMS – Data from XML documents are stored in a traditional
database – The DBMS is augmented with special-purpose sw for
managing XML data storage and retrieval • Native XML DBMS (NXDBMS)
– DBMS designed especially for storing XML documents
41
XML as a DBMS, Pros and Cons
+ Hierarchical model + Rich data structure
+ Unicode + Portability
+ Text format + Readable by humans
+ Schema language + XML Schema, DTD
+ Query languages + XPath, XQuery
+ APIs + SAX, DOM, JDOM
÷ Inefficient storage format ÷ Verbose ÷ Need to parse slow data
access ÷ Lacks
÷ indices ÷ transactions ÷ multiuser access ÷ security ÷ logging ÷ referential integrity ÷ ...
• Idea: XML documents are a natural format for storing data Use XML documents + XML tools + specialized sw as a ”DBMS”
42
XML as a DBMS, Conclusion
• Yes/maybe – small, single-user ”databases” (configuration
files, small contact lists,...) • In general: No
– must write a lot of code already found in modern DBMSs
– most production environments require a real DBMS
43
XML-Enabled DBMSs • For data-centric XML documents • When XML is used as a data transport format • How
– Store by shredding – Publish by composing
44
Properties of XML-Enabled DBMSs
• DB schema models primarily the data in the XML document – Data model is ”traditional”
• relational/object-relational/... – No XML ”visible” in DB
• Keeps existing data and applications intact – adding XML functionality is adding and
configuring data transfer (data mapping) sw
45
Properties of XML-Enabled DBMSs (cont.)
• XML mapping software – integrated into the DB engine or external to it
• Lossy modelling – XML document as such is discarded after shredding – In general, XML mapping sw can handle only a
subclass of XML documents • retains information that is important to the DB model:
data itself hierarchical relationships (parent, child, siblings) not entity references, CDATA sections, comments, processing
instructions, DTD, maybe not ordering of siblings,...
– Cannot reconstruct XML document in full detail
46
XML to DB Model Mapping • XML mapping characteristics
– Many-to-many • An XML document can be mapped to several DB schemas • Data of a DB schema can be mapped to several XML documents
– Design time • Runtime (automatic) mapping has to make too many assumptions
about DB schema
• Mapping techniques: – Bidirectional
• From XML documents to the DB schema, and vice versa – Unidirectional
• Using a query language to construct XML documents from a DB
47
XML to Relational DB Bidirectional Mapping Techniques
• Table-based – Restrictions on XML document structure
• Object-relational – XML document is viewed as a set of objects
48
Table-Based Mapping
• Restrictions on XML document structure – document structure must reflect RDB
table - row - column structure – Straightforward transformation between XML
schema and DB schema
49
Table-Based XML Document <Database> <SalesOrders> <SalesOrder> <Number>123</Number> <OrderDate>2003-07-28</OrderDate> <CustomerNumber>456</CustomerNumber> </SalesOrder> <SalesOrders> <Items> <Item> <Number>1</Number> <PartNumber>XY-47</PartNumber> <Quantity>14</Quantity> <Price>16.80</Price> </Item> <Item> <Number>2</Number> <PartNumber>B-987</PartNumber> <Quantity>6</Quantity> <Price>2.34</Price> </Item> </Items> </Database>
table
table
row
row
row
50
Resulting Table-Based XML and DB Schemas
DB schema (tables): SalesOrders (Number, OrderDate, CustomerNumber) Items (SONumber, Number, PartNumber, Quantity, Price) XML Schema: <!ELEMENT Database (SalesOrders, Items)> <!ELEMENT SalesOrders (SalesOrder*)> <!ELEMENT SalesOrder (Number, OrderDate, CustomerNumber)> <!ELEMENT Number (#PCDATA)> <!ELEMENT OrderDate (#PCDATA)> <!ELEMENT CustomerNumber (#PCDATA)> <!ELEMENT Items (Item*)> <!ELEMENT Item (SONumber, Number, PartNumber, Quantity, Price)> <!ELEMENT PartNumber (#PCDATA)> <!ELEMENT Quantity (#PCDATA)> <!ELEMENT Price (#PCDATA)>
51
Object-Relational Mapping
• XML document is viewed as a set of objects
• Use techniques from O-R to perform mapping – objects map to tables – object properties map to columns – inter-object relationships map to primary key/
foreign key relationships
53
Object-Relational Mapping <SalesOrder Number="123"> <OrderDate>2003-07-28</OrderDate> <CustomerNumber>456</CustomerNumber> <Item Number="1"> <PartNumber>XY-47</PartNumber> <Quantity>14</Quantity> <Price>16.80</Price> </Item> <Item Number="2"> <PartNumber>B-987</PartNumber> <Quantity>6</Quantity> <Price>2.34</Price> </Item> </SalesOrder>
item object
item object
SalesOrder object
55
XML to Relational DB Unidirectional Mapping Techniques • Mapping by querying
– SQL/XML • extending SQL with constructs for creating XML
fragments – XQuery
• for querying XML documents • requires a pre-mapping of the relational DB to one
or more virtual XML documents
56
SQL/XML
• Extensions to SQL for creating XML documents and document fragments from relational data – XML data type
• Valid XML data types: NULL, XML document, element content
– Scalar functions for creating XML • XMLELEMENT, XMLATTRIBUTES, XMLFOREST,
XMLCONCAT,...
57
Example SQL/XML Mapping
SQL/XML statement: SELECT XML2CLOB( XMLELEMENT( NAME Customer, XMLATTRIBUTES(customers.id AS ID), XMLELEMENT(NAME Name, customers.name) ) ) AS CustomerXML FROM customers
Result: For each row in customers, the following value in the CustomerXML column: <Customer ID="customer id"> <Name>customer name</Name> </Customer>
Schema: customers(id, name, address)
58
Example XQuery Mapping
Pre-mapped XML document: <customers> <Customer> <id>customer id</id> <name>customer name</name> <address>cust. address</address> </Customer> <Customer> .. </Customer> </customers>
XQuery statement: FOR $c IN document(”http://cust.xml")//Customer RETURN <Customer ID="{ $c/id }"> <Name>{ $c/name }</Name> </Customer>
Schema: customers(id, name, address)
59
NXDBMS: Native XML DBMS
• For document-centric XML documents – need to be able to retain XML documents
as-is • For semi-structured data
– when other DB models are inappropriate, e.g. since they require changing the DB schema when the XML schema changes/evolves
60
NXDBMS
• Specialized for storing XML data – Stores all components of the XML model
intact • XML document is the logical unit
– (Cf. the logical unit in an RDBMS, which is a row)
• Schema-independent NXDBMS: – Can store any XML document regardless of
its schema
61
Properties of NXDBMS
• Data model: – XML document is the fundamental (logical) unit of
storage – The NXDBMS defines a data model for XML
documents • As a minimum, elements, attributes, text (PCDATA),
document order • Ex.: XPath data model, XML Infoset, models implied by DOM
and the events in SAX 1.0
– In addition, the NXDBMS may support traditional DB models
62
Properties of NXDBMS (cont.)
• Implementation of own choice – as CLOBs – as fixed set of tables – as DOM trees in OODBs – as indexed set of hash tables
• Physical storage of own choice • NXDBMS abstracts away details;
application developer can build applications using XML technologies
63
Properties of NXDBMS (cont.) • Retrieval speed
– Can utilize physical storage if reflects XML document structure • Performance depends heavily on retrieval patterns coinciding with
XML document structure • Data export format
– XML (for almost all NXDBMSs) – If application needs other format, must parse
• cumbersome for local applications • OK for distributed applications that use XML as data transport
format • Schema-independent NXDBMSs
– Flexibility – Easy application development – Risk of low data integrity
64
Properties of NXDBMS (cont.) • Query languages
– XPath • Extended to query across collections of documents • Not designed as a database query language
– XQuery • Still under development
• Indices – What kind of indices are needed? – Along which of XML’s axes?
• child, descendant, parent, ancestor, following-sibling, preceding-sibling, following, preceding, attribute, namespace
• Updates – Currently hardly supported:
• Retrieve complete document, change, return to database
Conclusions
• Very large amounts of XML data and documents in the Internet -> Management?
• DBS only good for well-structured data • XML & DBS for semi-structured data
– 3 kinds of XDBMSs – Performance problems:
XML workload <-> DBS workload – Multi-layer hierarchical architectures
65
XML Overview • Similarities to semistructured models • Data is divided into pieces called elements • Elements can be nested, but not overlapped
– Nesting represents hierarchical relationships between the elements
• Elements have attributes • Elements can also have relationships to other elements
(ID-IDREF) • Can be represented as a graph, but usually simplified as
a tree – Root element – Zero or more child elements representing nested
subelements – Document order over elements – Attributes are also shown as nodes
68
XML Preliminaries
• XML – Extensible Markup Language – Data model – API – Schema languages – Query languages – Transformation languages
• Web services
69
XML – In a Nutshell Extensible Markup Language, World Wide Web Consortium (W3C)
• Standard of the W3C in 1998 • Documents (content) • XML Schema or DTD (structure) • Namespaces • XSL, XSLT (stylesheets, transformations) • XPath, XPointer, XLink • XQuery
• Tagged elements with attributes – Extensible, self-describing
• Tag names must be unique (use namespaces)
70
Namespaces • All tags must be defined in a namespace
– Unique global names using URI references
• W3C provides certain standard namespaces (xsd, xsi, html) • Organizations can provide domain specific namespaces
• Each document specifies the namespaces used in its schema – A default namespace - a standard or domain namespace – Its own target namespace - for elements defined in document – Others …
• Referencing a name (syntax) – Qualified names (namespace:name) xsd:element – name (default namespace implied) car
71
Elements • Element declarations (see XML Schema)
<xsd:element name="numStudents" type="xsd:positiveInteger"/> • Element definitions
– start tag <tagName> must be paired with an – end tag </tagName> – <emptyElement/>
<numStudents>31</numStudents> • Sub-elements may be repeated
– Subject to minOccurs and maxOccurs constraints (XML Schema) • Both default to 1 if not specified
• References <xsd.element name=“foo”, type=“xsd.string”/> … <xsd.element ref=“foo”/>
72
Attributes, Entities • Attributes (of an element)
– Name-value pairs placed in the start tags • name = “value” • Each name may only appear once (single-valued)
<elevation units="feet">5440</elevation> • Entity
– — an entity representing the character ‘—’ – starts with & and ends with ; – Used to distinguish reserved words from content
< represents the character ‘<‘ (also > & ' ")
– <!ENTITY myPic SYSTEM “me.jpg” NDATA JPEGFORMAT>
73
XML Document <? xml version=“1.0” ?> <racingTeams xmlns=“http://www.cart.org”
xmlns:xsi=“http://www.w3.org/2001/XMLSchema-Instance” xsi:schemaLocation=“http:// www.cart.org team.xsd”> <team> <name>“Shell</name> <car number=3><make>“Helix Ford”</make> <engine id=“47456”/></car> <car number=4><make>“Helix Ford”</make>…</car> <driver><name>“Steve Johnson”</name> <points>2217</points></driver > <driver><name>“Paul Radisich”</name></driver > <spares><engine id=“102555”/> <engine id=“102556”/></spares> </team> . . .
</racingTeams> <!-- Each document has one root element (e.g., racingTeams) whose name
matches a schema element or <!DOCTYPE name> -->
74
XML in Action 1. Export: XML document written to file or stream 2. Delivery: Passed to import software 3. Import: Parse XML document,
Validate using DTD or Schema (optional)
Schema or DTD
Application1 Application2 Document
Validate
Export Import
75
Semi-Structured Data • Semi-structured data is data that may be irregular or incomplete and
have a structure that may change rapidly or unpredictably. – It generally has some structure, but does not conform to a fixed
schema – Self-describing (e.g., in terms of XML element tags)
• Characteristics – Heterogeneous – Irregular structure – Large evolving schema – Hyperdata nodes and references
• Desirable to treat web sources like a database – Often XML documents are like semi-structured data
76
Semi-Structured Example DreamHome (&1) PropertyForRent (&5) Branch (&2) street (&11) “2 Manor Rd”
street (&7) “22 Deer Rd” type (&12) “Flat” Manager &3 monthlyRent (&13) 375
Staff (&3) OverseenBy &4 name (&8) PropertyForRent (&6) fName (&17) “John” street (&14) “18 Dale Rd” lName (&18) “White” type (&15) 1 ManagerOf &2 annualRent (&16) 7200
Staff (&4) OverseenBy &4 name (&9) “Ann Beech” salary (&10) 12000 Oversees &5 Oversees &6
77
XML and Data on the Internet • Primary purpose:
– Sharing of data across systems • In particular, across the Internet
– Standard to encode structured (and semistructured) information • Features:
– Text-based • Simultaneously human- and machine-readable format • Self-documenting format (structure & field names + data values)
– Can represent most general data structures: • records, lists, trees
– Platform-independent – Strict syntax – Simple, efficient, consistent parsing algorithms – Extensible
• Can create your own tags, i.e., your own vocabulary
78
XML Data Models • XML Information Set (Infoset)
– An XML document has an information set if it is well-formed and conforms to Namespaces standard
– XML document’s information set consists of information items with a set of named properties
– Usually created through parsing an XML document – Information set and information item similar in meaning to tree
and node respectively • but no one-to-one mapping with nodes of DOM or the tree and
nodes of the XPath data model – No requirements for specific interface
• Not only tree structure, also others, e.g., event-based, query-based – 11 different types of information items
79
XML API
• DOM – Interface-oriented API – Allows for navigation in the document as if it were a
tree – DOM parser creates a tree of objects before access is
allowed • SAX
– Lexical, event-driven interface – Document is read serially; contents reported as
”callbacks” to methods on a handler object
80
XML Schema Languages
• Purpose: to identify a class of XML documents – DTD – Document Type Definition – XML Schema
81
XML Schema • For declaring
– the organization of instance documents – the datatype of each element/attribute
• Features – XML-based syntax
• Can use XML tools to generate XML Schemas – Integrated with namespace mechanism
• xsd is standard default namespace for schemata • Included schema must have same targetNamespace
– Rich set of built-in types – Elements have nested scope – Keys definable in terms of elements – Referential integrity constraints enforced with keys – Options for ordered (sequences) or unordered (sets) – Validation of documents against a schema
82
XML Schema Grammar • Namespace • Compositors
<sequence>, <choice>, <all>, <union>, <list> • complexType, simpleType • complexContent, simpleContent • restriction, extension, enumeration • ID, IDREF [List(IDREF) replaces IDREFS ] • unique like a candidate key, whenever present => unique • key minOccurs > 0, nillable = “false”
– selector collection to which key applies (like Table) – field path expressions relative to selector collection
• keyref specifies foreign key constraint • Empty elements
83
XML Schema Example <?xml version="1.0"?> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema" targetNamespace="http://www.binary.org" xmlns="http://www.binary.org"> <xsd:element name="life"> <xsd:complexType> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element name="work" type="xsd:string"/> <xsd:element name="eat" type="xsd:string"/> </xsd: sequence> <xsd:choice> <xsd:element name="work" type="xsd:string"/> <xsd:element name="play" type="xsd:string"/> </xsd:choice> <xsd:element name="sleep" type="xsd:string"/> </xsd:sequence> </xsd:complexType> </xsd:element> </xsd:schema>
life ::= ( (work, eat)*, [ work | play ], sleep )*
84
Validating Your Data
<vehicles> <vehicle> <nickname>Count Zero</nickname> <model>Series I, 80"</model> <construction> <start>1949-07-21</start> <end>1949-08-09</end> </construction> </vehicle> </vehicles>
XML Schema
XML Schema validator Data is ok!
XML
Is this data valid? To be valid, it must meet these constraints (data business rules): 1. The vehicle must be comprised of a nickname, followed by a model, followed by a construction description. 2. Any number of vehicles can be listed in the contents of vehicles
85
XML Document <? xml version=“1.0” ?> <racingTeams xmlns=“http://www.cart.org”
xmlns:xsi=“http://www.w3.org/2001/XMLSchema-Instance” xsi:schemaLocation=“http:// www.cart.org team.xsd”> <team> <name>“Shell</name> <car number=3><make>“Helix Ford”</make><engine id=“247456”/></car> <car number=4><make>“Helix Ford”</make> <engine id=“102456”/></car> <driver><name>“Steve Johnson”</name> <points>2217</points></driver > <driver><name>“Paul Radisich”</name></driver > <spares> <engine id=“102555”/> <engine id=“102556”/> </spares> </team> . . .
</racingTeams> <!-- Each document has one root element (e.g., racingTeams) whose name
matches a schema element or <!DOCTYPE name> -->
87
XPath • View XML document as a tree of nodes and references (edges) • XPath: Sequence of steps that yields a set of nodes
/ step / step / step .... step = basis + [predicates] (Like a unix path, but each step can be a query expression)
/ root node . current node .. parent node @attr access attribute ”attr” [n] access the nth occurrence of a node at this level last() // descendent or self text() text content of node functions sum(), count(), ... [...] selection condition (predicate)
• Example:
//student[status=”ok” and starts_with(.//last, ”E”) and not (.//last=.//first)]
88
XQuery • W3C standard
– (XQuery 1.0) W3C recommendation Jan 2007 – Shares common datamodel with XPath 2.0 – Several implementations
• Data model can handle single documents, partial fragments, and collections of documents
• In same query can – specify what you are looking for – designate what its output format should look like
• Three language ”formats” – surface syntax – alternative XML-based syntax (XQueryX) – formal algebraic language
89
XQuery Expressions • Path expressions
– XPath syntax + dereference operation (->) + range predicate – Path expression returns an ordered list of nodes – Start at document(string), /, or //
• Element constructors • FLWR expressions
– FOR – LET – WHERE – RETURN
• Expressions involving operators and functions • Conditional expressions • Quantified expressions • List constructors • Expressions that test or modify data types
90
XQuery Expressions • Path expressions
– XPath syntax + dereference operation (->) + range predicate – Path expression returns an ordered list of nodes – Start at document(string), /, or //
• Element constructors • FLWOR expressions
– FOR – LET – WHERE – ORDER BY – RETURN
• Expressions involving operators and functions • Conditional expressions • Quantified expressions • List constructors • Expressions that test or modify datatypes
91
FLWOR (pronounced ”flower”)
• FOR iterates through a sequence of individual nodes out of the selected collection, in order, one at a time
• LET binds a variable to the set of nodes in the selected collection • FOR or LET clauses are considered nested; later clauses may
reference variables bound in previous clauses
FOR and LET clauses generate a list of tuples of bound expressions, preserving document order
WHERE clause applies a predicate, eliminating some of the tuples
RETURN clause is executed for each resulting tuple, generating an ordered list of outputs
ORDER BY clause rearranges the sequence of surviving tuples
FOR var IN expr
LET var IN expr WHERE expr ORDER BY expr RETURN expr
92
FLWR (2) • WHERE clause is optional
– Predicates using bound variables
• RETURN generates the output (constructor expression for output) – node | [ordered] forest of nodes | primitiveValue
– The constructor is executed once for each tuple of bindings from FOR or LET clauses that satifies the WHERE clause.
– It preserves ordering where ordering exists – The RETURN expression often contains Element contructors
• With references to bound variables and nested subelements
– A list may be constructed by enclosing zero or more expressions in square brackets, separated by commas
93
XQuery Example XML document: <?xml version="1.0" encoding="ISO-8859-1"?> <!-- Edited with XML Spy v2006 (http://www.altova.com) --> <bookstore> <book category="COOKING"> <title lang="en">Everyday Italian</title> <author>Giada De Laurentiis</author> <year>2005</year> <price>30.00</price> </book> <book category="WEB"> <title lang="en">XQuery Kick Start</title> <author>James McGovern</author> <author>Per Bothner</author> <author>Kurt Cagle</author> <author>James Linn</author> <author>Vaidyanathan Nagarajan</author> <year>2003</year> <price>49.99</price> </book> <book category="WEB"> <title lang="en">Learning XML</title> <author>Erik T. Ray</author> <year>2003</year> <price>39.95</price> </book> </bookstore>
Result: <title lang="en">Learning XML</title> <title lang="en">XQuery Kick Start</title>
XQuery: for $x in doc("books.xml")/bookstore/book where $x/price>30 order by $x/title return $x/title
94
XQuery Examples Path expression example: document("zoo.xml")/chapter[title = "Frogs"]//figref/@refid->fig/caption FLWR expression example: FOR $p IN distinct(document("bib.xml")//publisher) LET $a := avg(document("bib.xml") /book[publisher = $p]/price) RETURN <publisher> <name> $p/text() </name> <avgprice> $a </avgprice> </publisher>
Note: result(//) = result(.)+result(/) Probably here / and // give the same result
Note: this “publisher” name is in the local scope of document(“bib.xml”) from the context of the path expression And this “publisher” name must match the name of an element in the default namespace J
95
XML Transformation Languages • XSLT – Extensible Stylesheet Language
Transformations – Language for transforming an XML document into
another XML document, or to HTML, or another text-based format
– Rule-based • CSS – Cascading Style Sheets
– For describing the presentation (colors, fonts, layout,...) of a document written in a markup language
– Allows for separation of document content (written in e.g. XML) from document presentation (written in CSS)
– Non-XML notation
96
Example XSLT XSLT stylesheet <?xml version="1.0"?> <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
version="1.0"> <xsl:template match="/"> <HTML> <BODY> <TABLE BORDER="2">
<TR> <TD>Name</TD> <TD>Address</TD> <TD>Tel</TD> <TD>Fax</TD> <TD>Email</TD> </TR> <xsl:for-each select="PEOPLE/PERSON"> <TR> <TD><xsl:value-of select="NAME"/></TD> <TD><xsl:value-of select="ADDRESS"/></TD> <TD><xsl:value-of select="TEL"/></TD> <TD><xsl:value-of select="FAX"/></TD> <TD><xsl:value-of select="EMAIL"/></TD> </TR> </xsl:for-each>
</TABLE> </BODY> </HTML> </xsl:template> </xsl:stylesheet>
Resulting HTML putput <TABLE BORDER="2">
<TR> <TD>Name</TD> <TD>Address</TD> <TD>Tel</TD> <TD>Fax</TD> <TD>Email</TD> </TR> <TR> <TD>Mark Wilson</TD> <TD>911 Somewhere Circle, Canberra, Australia</TD> <TD>(++612) 12345</TD> <TD>(++612) 12345</TD> <TD>[email protected]</TD> </TR> ..
<TABLE>
Source XML file <?xml version="1.0" ?> <PEOPLE> <PERSON> <NAME>Mark Wilson</NAME> <ADDRESS>911 Somewhere Circle, Canberra, Australia</ADDRESS> <TEL>(++612) 12345</TEL> <FAX>(++612) 12345</FAX> <EMAIL>[email protected]</EMAIL> </PERSON> ... </PEOPLE>
97
Xpointer • Xpointer : an Xpath expression within a URI +
points (position within a document), or ranges (start-point, end-point)
– xpointer(id('list37')/item) – xpointer(id('list37')/item[1]/range-to(following-sibling::item[2]))
• http://someURL#xpointer(XPath expression) (entities, escapes)
– doc.xml#xpointer(string-range(//P,"a little hat ^^"))
• Nested elements from different namespaces (same local name)
– xmlns(x=http://example.com/foo) xmlns(y=http://example.org/bar) xpointer(//x:a/y:a)
98
XML Metadata Interchange • XMI is an XML language to interchange the objects of software
systems: – UML/MOF1 designs – Computer software classes and interfaces – Databases and database schemas – DTDs and XML schemas
• Added to the list of OMG adopted technologies in February, 1999 as XMI 1.0 • Most recent minor revision is XMI 1.1 (November 2000)
1 MOF = Meta-Object Facility, describing data about data, i.e., the classes and associations structure of the data
99
XMI: Sharing Objects • XML - Sharing Data • XMI - Sharing Objects
– Creates custom • XML Schemas (DTDs) • XML documents
from class definitions.
Application1
Objects
Data
Text
Application2
Objects
Data
Text
XMI
XML
Unicode
Key
100
XMI Object Mapping
MOF Concept XML Concept • Class XML element, odd levels
• Object typed attribute XML element, even levels
• Containment XML element, even levels
• Reference XML attribute IDREF
• Basic attribute (int, String) XML attribute CDATA
101
XMI Mapping Example Class Car { // Class
String color; // Field, String Person customer; // Field, Reference Class Definition Engine standardEngine; // Field, Object Engine optionalEngine; // Field, Object
}
<Car color="Blue" customer="Joe"> <Car.standardEngine> <Engine ..... /> XMI Instance </Car.standardEngine> <Car.optionalEngine> <Engine ..... /> � XML element, odd levels </Car.optionalEngine> � XML element, even levels
</Car> � XML attribute IDREF � XML attribute CDATA
102
Validation in XMI
MOF model (XMI Document)
Generated XML Schema
Model
Objects (XMI Document) XML Parser
Instance
Well-formed
Generate
Semantic Validation Optional Syntactic Validation
103
XML Schema: Specifies the Properties for a Class of Resources
<xsd:complexType name="Book"> <xsd:sequence> <xsd:element name="Title" type="xsd:string"/> <xsd:element name="Author" type="xsd:string" maxOccurs="unbounded"/> <xsd:element name="Date" type="xsd:year"/> <xsd:element name="ISBN" type="xsd:string"/> <xsd:element name="Publisher" type="xsd:string"/> </xsd:sequence> </xsd:complexType>
"For the class of Book resources, we identify five properties - Title, Author, Date, ISBN, and Publisher"
104
XML Instance Document: Specifies Values for the Properties
<Book> <Title>My Life and Times</Title> <Author>Paul McCartney</Author> <Date>July, 1998</Date> <ISBN>94303-12021-43892</ISBN> <Publisher>McMillin Publishing</Publisher> </Book>
"For a specific instance of a Book resource, here are the values for the properties.
Use schemaLocation to identify the companion document (i.e., the schema) which defines the Book class of resources."
105
Metadata Interchange using XML Schema
• XML Schema Strategy - two documents are used to provide metadata: – a schema document specifies the properties (metadata) for a
class of resources (objects); – an instance document provides specific values for the properties
of each object instance being transferred.
• Describing a schema for an object model using XML Schema requires another XML Schema that describes the object model’s metamodel!
• XMI uses an XML Schema for the UML metamodel J